Content uploaded by Elisabete Barbarino
Author content
All content in this area was uploaded by Elisabete Barbarino
Content may be subject to copyright.
Journal of Applied Phycology (2005) 17: 447–460
DOI: 10.1007/s10811-005-1641-4 C
Springer 2005
An evaluation of methods for extraction and quantification of protein from
marine macro- and microalgae
Elisabete Barbarino1,2& Sergio O. Louren¸co2,∗
1Programa de P´
os-Graduac¸ ˜
ao em Biotecnologia Vegetal, Universidade Federal do Rio de Janeiro, Rio de Janeiro,
RJ, Brazil; 2Departamento de Biologia Marinha, Universidade Federal Fluminense, Caixa Postal 100644, CEP
24001-970, Niter´
oi, RJ, Brazil
(∗Author for correspondence: e-mail: solourenco@yahoo.com, fax: +5521-2629-2292)
Received 19 December 2004; accepted 27 June 2005
Key words: amino acids, marine macroalgae, marine microalgae, nitrogen, protein determination, seaweeds
Abstract
Comparison of data of protein content in algae is very difficult, primarily due to differences in the analytical meth-
ods employed. The different extraction procedures (exposure to water, grinding, etc.), protein precipitation using
different amounts of 25% trichloroacetic acid and quantification of protein by two different methods and using
two protein standards were evaluated. All procedures were tested using freeze-dried samples of three macroalgae:
Porphyra acanthophora var. acanthophora,Sargassum vulgare and Ulva fasciata. Based on these results, a protocol
for protein extraction was developed, involving the immersion of samples in 4.0mL ultra-pure water for 12 h, fol-
lowed by complete grinding of the samples with a Potter homogeniser. The precipitation of protein should be done
with 2.5:1 25% TCA:homogenate (v/v). The protocol for extraction and precipitation of protein developed in this
study was tested with other macroalgae (Aglaothamnion uruguayense,Caulerpa fastigiata,Chnoospora minima,
Codium decorticatum,Dictyota menstrualis,Padina gymnospora and Pterocladiella capillacea)and microalgae
(Amphidinium carterae,Dunaliella tertiolecta,Hillea sp., Isochrysis galbana and Skeletonema costatum). Compar-
ison with the actual protein content determined from the sum of amino acid residues, suggests that Lowry’s method
should be used instead of Bradford’s using bovine serum albumin (BSA) as protein standard instead of casein. This
may be related to the reactivity of the protein standards and the greater similarity in the amino acid composition of
BSA and algae. The current results should contribute to more accurate protein determinations in marine algae.
Introduction
Determination of protein content of algae can provide
important information on the chemical characteristics
of algal biomass. The methods most commonly used
to quantify protein are: (i) the alkaline copper method
(Lowry et al., 1951); (ii) the Coomassie Brilliant Blue
dye method (Bradford, 1976); or (iii) determination of
crude protein (N×6.25).
The calculation of protein content by N×6.25 re-
quires some caution, not always considered by authors
using this method. Plant materials, fungi and algae
commonly have high concentrations of non-protein ni-
trogenaceous substances such as pigments (chlorophyll
and phycoerythrin), nucleic acids, free amino acids
and inorganic nitrogen (nitrate, nitrite and ammonia)
(Louren¸co et al., 1998; Conklin-Brittain et al., 1999;
Fujihara et al., 2001) whose presence makes the factor
6.25 unsuitable since it overestimates the actual pro-
tein content (Ezeagu et al., 2002). Specific nitrogen-
to-protein conversion factors were recently proposed
for 12 marine microalgae (Louren¸co et al., 2004) and
19 seaweeds (Louren¸co et al., 2002), varying from 3.75
for Cryptonemia seminervis ared alga, to 5.72 for Pad-
ina gymnospora abrown alga.
The determination of protein by the Lowry and
Bradford methods is carried out by spectrophotome-
try. The Lowry method detects protein by a reaction
catalyzed by copper, a component of the Folin phe-
nol reactions. The chemical reaction detects peptide
448
bonds and is also sensitive to some amino acids such
as tyrosine and tryptophan (Legler et al., 1985). In the
Bradford method, the Coomassie Brilliant Blue dye
is bound to protein mainly by arginine residues and
to a lower degree by histidine, lysine, tyrosine, tryp-
tophan and phenylalanine residues. The binding be-
tween the dye and amino acids is attributed to van
der Waals forces and hydrophobic interactions (Comp-
ton & Jones, 1985). As a consequence, the reactiv-
ity of both methods in comparison to a specific pro-
tein is strongly influenced by its amino acid compo-
sition, since not all amino acids can oxidate equally
the Folin phenol reactive or bind to the Coomassie dye
(Stoscheck, 1990). The differences in the principles
of the methods contribute to making comparison of
results available in the literature even more difficult,
since the choice of the method to be used is an arbitrary
decision.
Bovine serum albumin (BSA) is the most used pro-
tein standard for calibration curves in spectrophotome-
try, but many other proteins can be used. Several studies
suggest that the Lowry and Bradford analyses produce
different measurements of protein when using BSA as
the protein standard for samples such as the gut fluid
of fish (Crossman et al., 2000), marine invertebrates
(Zamer et al., 1989), higher plants (Eze & Dumbroff,
1982) and marine phytoplankton (Clayton et al., 1988).
To obtain a more reliable measurement of protein, it
would be useful to identify the predominant proteins
in the cells (Berges et al., 1993). However, this rec-
ommendation has no practical value, considering the
difficulty of extracting, purifying and characterising
the main proteins present in the cells and subsequently
using them as protein standards. Nguyen & Harvey
(1994) suggested the use of ribulose-1,5-diphosphate
carboxylase (RuDPCase) for calibration curves to anal-
yse samples of photosynthetic organisms, since RuD-
PCase corresponds to about 15% of the total protein in
chloroplasts.
Several substances may interfere with both the
Lowry and Bradford method, such as phenol and phe-
nolases (Mattoo et al., 1987), glucosamine and de-
tergents (Peterson, 1979) and flavonoids (Compton &
Jones, 1985) among many others (see the comprehen-
sive studies of Peterson, 1979; Stoscheck, 1990 on in-
terfering substances). These substances could affect
analyses by either increasing the absorbance (overesti-
mating values), or decreasing the measurements by in-
hibiting the action of specific reagents. However, their
influence may be avoided by precipitation of the pro-
tein sample with trichloroacetic acid (TCA). Concen-
trations of TCA between 0.18 and 0.34 M can be used
to seperate protein from the other extract components,
because only protein is precipitated (Clayton et al.,
1988). The physical separation among protein, small
peptides and free amino acids is especially important,
since the analytical methods are sensitive to the last two
classes of substances. Thus, protein precipitation with
TCA is strongly recommended to avoid the quantifica-
tion of small peptides and free amino acids (Nguyen
& Harvey, 1994) as well as interference by other
substances.
As many variables are simultaneously involved with
protein analysis, the influence of specific factors may
be neglected by authors, affecting the accuracy of pro-
tein analysis. However, studies focussing on protein
analysis in algae are relatively uncommon and exper-
imental data are needed to fill this gap. It is also very
important to develop a simple and inexpensive pro-
tocol, using low-cost equipment and consumables, in
order to make it accessible to everyone interested in
data on algal protein: researchers, algae producers, peo-
ple interested on the nutritional value of algae and so
on.
In this study, different procedures for extraction and
quantification of the protein content of marine algae
were evaluated. The specific aims of this study were: (i)
to create a protocol for the extraction and quantification
of protein of marine algae; (ii) to compare the use of two
methods (Lowry and Bradford) for the determination
of protein in marine algae; (iii) to evaluate the effects
of the time of extraction, the use of grinding and the
precipitation of samples with TCA on the quantity of
protein extracted; and (iv) to compare the amino acid
profile of the algal samples and the protein standards
(BSA and casein) used.
Materials and methods
Fifteen species of marine algae covering a wide taxo-
nomical range were analysed. The field-collected ma-
rine were identified following the checklist of Wynne
(1998). Marine microalgae were cultured in the labora-
tory. The classification below is based on Lee (1999):
Chlorophyta
1. Chlorophyceae: Dunaliella tertiolecta Butcher
(Volvocales).
2. Ulvophyceae: Caulerpa fastigiata Montagne
(Bryopsidales), Codium decorticatum (Woodw.) M.
Howe; (Bryopsidales) and Ulva fasciata Delile
(Ulvales).
449
Cryptophyta
Cryptophyceae: Hillea sp. Schiller (Cryptomon-
adales).
Dinophyta
Dinophyceae: Amphidinium carterae Hulburt
(Gymnodiniales).
Heterokontophyta
Bacillariophyceae: Skeletonema costatum (Gre-
ville) Cleve (Biddulphiales).
Phaeophyceae: Chnoospora minima (K. Hering)
Papenfuss (Scytosiphonales), Dictyota menstrualis
(Hoyt) Schnetter, H¨ornig et Weber-Peukert (Dic-
tyotales), Padina gymnospora (K¨utzing) Sonder
(Dictyotales) and Sargassum vulgare C. Agardh
(Fucales).
Prymnesiophyta
Prymnesiophyceae: Isochrysis galbana Parke
(Pavlovales).
Rhodophyta
Bangiophycidae: Porphyra acanthophora var.
acanthophora E. C. Oliveira and Coll (Bangiales).
Florideophycidae: Aglaothamnion uruguayense
(Taylor) Aponte, Ballantine et Norris (Ceramiales)
and Pterocladiella capillacea (S. G. Gmel.) San-
telices et Hommersand (Gelidiales).
All species of marine macroalgae were collected
in June 1998 at Rasa Beach (located in Arma¸c˜ao de
B´uzios, 22◦44S and 41◦57W) and Per´o Beach (lo-
cated in Cabo Frio, 22◦51S and 41◦58W), Northern
Rio de Janeiro State, Brazil. Whole thalli of adult plants
were collected early in the morning and washed in the
field with seawater to remove epiphytes, sediment and
organic matter. Algae were packed in plastic bags and
kept on ice until returned to the laboratory. In the lab-
oratory, samples were gently brushed under running
seawater, rinsed with distilled water, dried with paper
tissue, frozen at −20 ◦C and freeze-dried. The dried
material was powdered manually with the use of mor-
tar and pestle and kept in desiccators containing silica-
gel and protected from light at room temperature until
chemical analysis.
Culture of microalgae
All microalgal strains used in this study are available
at the Elizabeth Aidar Microalgae Culture Collection,
Department of Marine Biology, Federal Fluminense
University, Brazil. Starter cultures of 50–100 mL in
mid-exponential growth phase were inoculated into
2.0 L of seawater, previously autoclaved at 121◦C for
30 min in 3.0 L borosilicate flasks, and enriched with
Conway nutrient solution (Walne, 1966). Each experi-
ment was carried out in four culture flasks, exposed to
300 µmol photons m−2s−1(measured with a Biospher-
ical Instruments quanta meter QLS100) from beneath,
provided by fluorescent lamps (Sylvania daylight
tubes), under a 12:12 h light:dark cycle. Mean temper-
atures were 23 ±1◦Cinthe light period and 20 ±1◦C
in the dark period. Salinity of the culture medium was
32.0‰. Growth rates were calculated daily by direct
microscopic cell counting with Fuchs–Rosenthal or
Malassez chambers. Cultures were bubbled with fil-
tered air at a rate of 2 L min−1. The culture medium
was not buffered and pH was determined daily.
Each culture was sampled in the stationary growth
phase only. Cultures were concentrated by centrifuga-
tion at 7000 gfor 10min at 15 ◦C, at least once. Before
the last centrifugation, cells were washed with artificial
seawater (Kester et al., 1967) prepared without nitro-
gen, phosphorus and vitamins and adjusted to 15‰
salinity to remove any residual nitrogen from the cul-
ture medium. All supernatants obtained for each sam-
ple were combined and the cell number was determined
in this pool to quantify possible cell losses. The pel-
lets were frozen at −20 ◦C, freeze-dried (as described
above), weighed and stored in desiccators under vac-
uum and protected from light at room temperature until
analysis was done.
Amino acid analysis
Samples containing 5.0 mg of protein were acid hy-
drolysed with 1.0 mL of 6 N HCl in vacuum-sealed
hydrolysis vials at 110 ◦C for 22 h. Norleucine was
added to the HCl as an internal standard. Although
tryptophan was completely lost with acid hydrolysis
and methionine and cystine +cysteine could be de-
stroyed to varying degrees by this procedure, the hy-
drolysates were suitable for analysis of all other amino
acids. The tubes were cooled after hydrolysis, opened,
and placed in a descicator containing NaOH pellets
under vacuum until dry (5–6 days). The residue was
then dissolved in a suitable volume of a sample di-
lution Na–S R
buffer (Beckman Instr.), pH 2.2, filtered
through a Millipore membrane (0.22 µm pore size) and
analysed for amino acids by ion-exchange chromatog-
raphy in a Beckman, model 7300 instrument equipped
with an automatic integrator. Ammonia content is also
presented as it comes from the degradation of some
amino acids (e.g. glutamine, asparagine) during acid
hydrolysis (Moss´e, 1990).
450
Total nitrogen
Total nitrogen (TN) content was determined by CHN
analysis. 0.8–1.5 mg freeze-dried samples were com-
busted in a CHN analyser (Perkin–Elmer, model 2400).
Helium was used as carrier gas. Acetanilide (C=
71.09%; N=10.36%; H=6.71%) and/or benzoic
acid (C=68.84%; H=4.95%) were used to calibrate
the instrument.
Extraction of protein
Eight procedures for protein extraction were tested in
this study. All procedures start with 50mg of freeze-
dried algal sample, ground manually with pestle and
mortar. Two different volumes of water were tested
(1.0 and 4.0 mL), as well as two different incubation
periods of samples with water (6 and 12 h). In all the
cases samples were kept at 4 ◦C during the incubation
period.
In four out of the eight extraction procedures, sam-
ples were also ground using a Potter homogeniser (Mar-
coni, model MA099) after the incubation with water.
Samples were water-ground with a glass pestle and a
teflon mortar at medium speed. During grinding, sam-
ples were kept cool by the use of a circulating cooling
bath through the pestle. The grinding of samples was
started 1 h before the end of incubation of the sam-
ples with water. Six replicates were prepared for each
treatment and each species. Procedures for protein ex-
traction are based on Fleurence et al. (1995) with some
modifications (e.g. speed of centrifugation, grinding of
the samples, time of incubation). The specific proce-
dures for protein extraction analysed in this study are
as follows:
Procedure I. Algal samples were immersed in 1 mL of
ultra-pure water for 12 h. After the incubation pe-
riod, suspensions were centrifuged at 4 ◦C, 15,000 g
for 20 min. Supernatants were collected for protein
assay and the pellets re-extracted with 1.0 mL 0.1 N
NaOH with 0.5% β-mercaptoethanol (v/v). The mix-
ture of NaOH and pellets were kept at room tempera-
ture for 1 h with occasional manual shaking and then
centrifuged at 21 ◦C, 15,000 gfor 20 min. The sec-
ond supernatants were combined with the first ones
and the pellets were discarded. The final volume of
the extract was 2.0 mL.
Procedure II. Similar to procedure I, with one addi-
tional step included: the grinding of samples with a
Potter tissue homogeniser for 5 min, 1 h before the
end of the incubation period. Seven millilitre of ultra-
pure water was added to the system to rinse the Potter
homogeniser after grinding each sample to recover
all water-ground material. After this step, samples
were treated as described after the end of the incu-
bation period for procedure I. The final volume of
the extract was 9.0 mL.
Procedure III. This treatment is similar to procedure
I, differing by the use of 4 mL of water to incubate
dried samples, instead of 1 mL. The final volume of
the extract was 5.0 mL.
Procedure IV. This treatment is similar to procedure II,
differing by the use of 4 mL of water for incubating
dried samples, instead of 1 mL. Four millilitre of
ultra-pure water were added to the system to rinse
the Potter homogeniser after grinding each sample
to recover all water-ground material. After this step,
samples were centrifuged as described for procedure
I. The final volume of the extract was 9.0mL.
Procedure V. Similar to procedure I, differing only due
the incubation period: 6 h instead of 12 h. The final
volume of the extract was 2.0mL.
Procedure VI. Dried samples are treated such as in pro-
cedure II, except by the shorter incubation period: 6 h
instead of 12 h. The final volume of the extract was
9.0 mL.
Procedure VII. Similar to procedure III, differing only
due to the shorter incubation period: 6 h instead of
12 h. The final volume of the extract was 5.0mL.
Procedure VIII. Samples were treated as described in
procedure IV, differing only by the shorter incuba-
tion period: 6 h instead of 12 h. The final volume of
the extract was 9.0 mL.
Asummary of the procedures to extract algal protein
is shown in Table 1.
Table 1. Summary of the procedures for extracting protein used in
this study
Volume of Period of
Treatment water (ml) incubation (h) Grinding
Procedure I 1.0 12 No
Procedure II 1.0 12 Yes
Procedure III 4.0 12 No
Procedure IV 4.0 12 Yes
Procedure V 1.0 6 No
Procedure VI 1.0 6 Yes
Procedure VII 4.0 6 No
Procedure VIII 4.0 6 Yes
451
The use of 1.0 mL 0.1 N NaOH with 0.5% β-
mercaptoethanol (v/v) for re-extracting the pellets was
adopted in all the eight procedures tested, independent
of the time of incubation and volume of ultra-pure wa-
ter used to incubate the samples.
Precipitation of protein
Protein precipitation was followed Berges et al. (1993).
Two proportions of cold 25% trichloroacetic acid
(TCA) (4 ◦C) added to the extracts were tested: 2.5:1
and 3.0:1 (TCA:homogenate, v/v). Tubes contain-
ing TCA and homogenate were kept in an ice bath
for 30 min and then centrifuged for 20 min at 4 ◦C
(15,000 g). Supernatants were discarded, pellets were
washed with cold 10% TCA (4 ◦C) and centrifuged
again. Pellets formed after the second centrifugation
were suspended in 5% TCA at room temperature, in a
proportion of 5:1 (5% TCA:precipitate, v/v) and cen-
trifuged at 21 ◦C (15,000 g) for 20 min. Supernatants
were discarded and pellets were kept in the tubes un-
til quantification of protein was done a few minutes
later. When the protein analysis was not performed im-
mediately, pellets were stored at −20 ◦C until further
analysis, following Dortch et al. (1984).
Precipitated protein was suspended in 0.5 mL 1.0N
NaOH and 2.0 mL 0.1N NaOH for the a Bradford and
Lowry assays, respectively. Aliquots were also col-
lected from the crude extracts obtained before the pre-
cipitation with TCA to perform protein analysis by the
Lowry method without previous precipitation.
Protein analysis
In the Lowry method, the Folin–Ciocalteu reactive
(Folin & Ciocalteu, 1927) (Sigma Co.) was diluted in
two volumes of ultra-pure water (1:2) and 0.5mL of
the diluted reactive was added to 1.0mL of sample,
previously mixed with 5.0mL of the reactive “C” [50
volumes of reactive “A” (2.0% Na2CO3+0.1 N NaOH)
+1volume of reactive “B” (1/2 volume of 0.5% CuSO4
5H2O+1/2 volume of 1.0% C4H4NaO64H2O)]. After
the addition of each reactive, samples were stirred for
2sin a test tube stirrer. Absorbance was measured at
750 nm, 35 min after the start of the chemical reaction
at room temperature.
In the Bradford assay, the Coomassie Brilliant Blue
dye G-250 (CBBG) binds to the protein. The bind-
ing of the dye with the protein is very quick and the
protein-dye complex remains soluble for 1 h. One hun-
dred milligram of CBBG (Sigma Co.) was dissolved in
50 mL 95% ethanol (Merck Co.) with a further addi-
tion of 100 mL 85% H3PO4(Merck Co.). The solution
was diluted with ultra-pure water to 1.0 L. Five millil-
itre of the reactive was used for each 0.1mL sample.
Absorbance was measured at 595 nm, 5 min after the
start of the chemical reaction at room temperature.
Calibration curves were prepared using bovine
serum albumin (BSA) (Sigma Co.) and casein (Sigma
Co.) at maximum concentrations of 100 µgmL
−1
(Lowry method) and 100 µg 0.1 mL−1(Bradford
method). Casein was diluted in ultra-pure water plus
some drops of 0.1 N NaOH. All measurements were
done using a Shimadzu, model UV Mini 1240 spec-
trophotometer.
In addition to colorimetric assays of protein, crude
protein for each species was also calculated using spe-
cific nitrogen-to-protein conversion factors proposed
by Louren¸co et al. (2002, 2004) as follows: A. carterae
(5.13), A. uruguayense (3.94), C. decorticatum (5.34),
C. fastigiata (4.52), C. minima (5.70), D. menstrualis
(4.55), D. tertiolecta (4.39), Hillea sp. (4.93), I. gal-
bana (5.07), P. acanthophora var. acantophora (4.47),
P. capillacea (4.78), P. gymnospora (5.72), S. costatum
(4.53), S. vulgare (5.53), and U. fasciata (5.59).
Statistical analysis
The results were analysed by one-way analysis of vari-
ance (ANOVA) with significance level α=0.05 (Zar,
1996) followed, where applicable, with Tukey’s mul-
tiple comparison test. In some cases, Student’s t-test
was used instead of ANOVA when comparing only two
treatments for each variable.
Results
Tests of protein extraction
The use of the Potter homogeniser produced remark-
able differences in the extraction of protein for the three
species tested. In all cases, significantly (p<0.001)
higher concentrations of protein was obtained in the
treatment with the use of the Potter homogeniser (Fig-
ures 1A–C). For Sargassum vulgare (Figure 1B) dif-
ferences in values obtained for samples extracted with
and without the Potter homogeniser were about 50%.
Higher values of protein were also obtained for samples
of Porphyra acanthophora var. acanthophora and Ulva
fasciata when extracted with the Potter homogeniser.
452
Figure 1. Quantification of protein in three species of marine
macroalgae (A) Porphyra acanthophora var. acanthophora, (B) Sar-
gassum vulgare and (C) Ulva fasciata by the Lowry (Lwy) and Brad-
ford (Bdf) methods, using bovine serum albumin (BSA) and casein
(CAS) as protein standards in the calibration curves. Three vari-
ables evaluated: (i) the water volume used to incubate samples (1
and 4 mL); (ii) the length of the incubation period for the extraction
of protein (6 and 12 h); and (iii) the use of a Potter homogenizer
for grinding samples. All assays included the precipitation of pro-
tein with 25% TCA, in a proportion of 2.5:1 (TCA:homogenate).
Mean ±S.D.(n=6).
Higher concentrations of protein were obtained
when 4.0 mL of water was used for U. fasciata (p<
0.001) (Figure 1C), and a longer period of incubation
(12 h) for S. vulgare (p=0.002) (Figure 1B). There
were no differences for P. acanthophora var. acan-
thophora (Figure 1A) incubated in wither 1.0 or 4.0 mL
for 6 h (p=0.52).
The precipitation of protein
No differences in the precipitation of protein were ob-
served for the two TCA:homogenate ratios using the
Lowry (0.12 ≤p≤0.70) and Bradford (0.14 ≤p≤
0.97) methods.
The quantification of protein in the tests
The results show large differences for all the three
species between the two protein quantification meth-
ods; values obtained with the Lowry method were al-
ways higher in all comparisons evaluated (p<0.01)
(Figures 1A–C).
The use of BSA in calibration curves seems to gen-
erate higher values in the samples of P. acanthophora
var. acanthophora (p<0.02). However for U. fasci-
ata, using the Lowry method (p=0.56) and S. vulgare
with the Bradford method (p=0.06), no differences
were found when comparing results with BSA or casein
as calibration standards.
In addition to the general analyses described above,
some tests on samples spiked with protein were per-
formed to assess any possible effects of endogenous
algal proteases in the samples which could partially de-
stroy protein during incubation (P´erez-Llor´enz et al.,
2003). Tests were carried out for all the three algae at
same dilutions used to extract and quantify algal pro-
tein. Controlled amounts of BSA were used as an inter-
nal standard: 2.5 mg of BSA were added to each flask in
which the algal material was incubated, such as in pro-
cedure IV. Controls included incubation of BSA with
no algae, following the same steps described for algal
samples and direct dilution and measurement of BSA,
without incubation. The results (data not presented)
showed no loss of protein after the incubation period
in all experiments carried out (p≥0.083), indicating
no activity of algal proteases at 4 ◦C, the incubation
temperature used.
Protocol for extraction and precipitation of protein
Results suggest the use of a general protocol involving
extraction of protein from algal samples using 4.0 mL
of ultra-pure water, for 12h and grinding of samples
with a Potter homogeniser. Samples should be precipi-
tated with TCA:homogenate (2.5:1 v/v). A diagram of
the protocol is shown in Figure 2. This protocol was
used for protein determination of the other algal species
using both the Bradford and Lowry methods, using
453
Figure 2.Diagrammatic representation of the protocol for extraction and precipitation of algal protein developed in this study.
454
BSA and casein as protein standards in the calibration
curves.
Amino acid profile of the algae and protein standards
The comparison of the amino acid profile of the algal
species (Table 2) shows great differences for aspartic
acid, glutamic acid and arginine. On the other hand,
some amino acids, such as glycine and leucine, had
similar values for the 15 species studied.
The comparison of the algal amino acid profile with
the protein standards show that the concentration of
glutamic acid in casein, and lysine in BSA, are higher
than those reported for all the algae. Both protein stan-
dards were lower in methionine compared to the algae.
Casein and BSA also show remarkable differences with
each other regarding some amino acids, such as alanine
(higher concentration in BSA) and proline (higher con-
centration in casein).
Quantification of protein
The highest percent of protein was measured in the red
algae A. uruguayense (15.6±0.3% of the dry matter),
followed by the cryptomonad Hillea sp. (15.3±0.6%)
(Table 3). The other microalgae had similar protein
contents, varying from 11.4±1.0%. (D. tertiolecta)
to 10.1±0.8% (I. galbana). The edible red algae P.
acanthophora var. acanthophora had 8.9±0.7% of
protein and P. capillacea had the lowest protein content
among the red algae (4.2±0.3%). In the brown algae,
small variations in protein content were found (8.7,
7.8 and 6.9% for P. gymnospora,C. minima and S.
vulgare, respectively), except for D. menstrualis, which
had a lower protein content (4.0±0.2%). The green
macroalgae had protein contents varying from 5.5%
(C. fastigiata)to7.3% (U. fasciata)(Table 3).
Confirming the same trend obtained with the ini-
tial results for the three macroalgae (P. acanthophora
var. acanthophora,S. vulgare and U. fasciata), all other
species gave significantly higher values of protein when
quantified using the Lowry method compared to the
Bradford method. In some cases (e.g. C. fastigiata), dif-
ferences between mean values were higher than 50%
(Table 3). For the microalgae, the variation between
the two methods was less, varying from 1.5 (I. gal-
bana, BSA as protein standard) to 1.9 (A. carterae,
BSA and casein as protein standard). For the macroal-
gae, the Lowry:Bradford ratio varied from 1.5 (D. men-
strualis, BSA and casein as protein standard) to 3.2 (C.
fastigiata, casein as protein standard). For all species,
values obtained with BSA as the protein standard were
higher than those obtained with casein (Table 3). Crude
extracts analysed using the Lowry method always gave
higher values than those obtained with precipitated
samples (p<0.01), except for A. uruguayense and
Hillea sp. (Table 3).
The 15 algae species showed great differences re-
garding total nitrogen and crude protein (Table 3). Mi-
croalgae had smaller variations in the total N,vary-
ing from 3.35% (I. galbana)to4.69% (A. carterae).
Variations in the total Nwere greater in the macroal-
gae, ranging from 1.94% (C. minima)to5.68% (A.
uruguayense). The calculations of crude protein gave
wide variations among species, varying from 10.52%
in C. decorticatum to 25.04% in Hillea sp. (Table 3).
The sum of amino acid residues varied from 9.99% (C.
minima)to20.11% (Hillea sp.). Microalgae tended to
have higher percentages of amino acid residues than
macroalgae.
Discussion
Extraction and precipitation of algal protein
Protein content of the three main algae species tested in
this study varied greatly depending on the different ex-
traction procedures tested. The efficiency of extraction
seems to be influenced directly by two main factors:
the chemical composition of the species and its mor-
phological and structural characteristics. The chemi-
cal composition of the three species is very distinct
(Louren¸co et al., 2002) and, in theory, this may lead
to differences in the protein content. S. vulgare is a
branched algae and possesses a hard and leathery thal-
lus, while P. acanthophora var. acanthophora and U.
fasciata have flattened soft thalli.
In the present study, the effects of lyophylisation on
the thalli should be also considered since freeze-dried
samples tend to be more difficult to extract, especially
leathery species such as C. minima and S. vulgare. This
means that better preservation by freeze-drying makes
them more difficult for further protein extraction. This
problem can be solved by grinding samples with Potter
homogeniser.
For P. acanthophora var. acanthophora lyophilisa-
tion, the main factor influencing the yield in the pro-
tein extraction was the use of grinding. This is prob-
ably related to the thallus form of this species. Ulva
fasciata has the same kind of thallus, but the use of
455
Table 2. Amino acid profile of 15 species of marine algae and two standards of pure protein: bovine serum albumin (BSA) and caseina
Protein standards Chlorophyta Cryptophyta Dinophyta
Amino acid BSA Casein C. fastigiata C. decorticatum D. tertiolecta U. fasciata Hillea sp. A. carterae
Cisteic acid 4.5 ±0.4 0.9 ±0.5 2.8 ±0.3 1.1 ±0.3 0.7 ±0.2 0.5 ±0.1 0.8 ±0.2 0.2 ±0.0
Aspartic acid 9.6 ±0.2 6.4 ±0.3 8.8 ±0.8 10.7 ±0.1 12.3 ±0.5 13.4 ±1.1 13.1 ±0.2 9.1 ±0.2
Treonine 4.8 ±0.2 3.7 ±0.2 4.7 ±0.5 6.0 ±0.1 4.6 ±0.2 5.2 ±0.2 4.6±0.0 5.1 ±0.1
Serine 3.8 ±0.4 4.8 ±0.3 6.0 ±0.6 5.0 ±0.2 3.6 ±0.1 5.9 ±0.7 3.8 ±0.0 5.5 ±0.3
Glutamic acid 16.7 ±0.3 21.1 ±0.8 10.4 ±1.1 12.0 ±0.9 12.8 ±0.4 12.9 ±0.4 12.1 ±0.0 13.6 ±0.3
Proline 4.7 ±0.3 11.3 ±0.6 5.1 ±0.4 4.8 ±0.7 4.9 ±0.2 4.7 ±0.1 3.5±0.0 4.2 ±0.3
Glycine 1.7 ±0.3 1.6 ±0.2 6.9 ±0.8 7.2 ±0.5 5.8 ±0.1 6.7 ±0.2 7.2±0.5 5.1 ±0.2
Alanine 5.4 ±0.2 2.6 ±0.3 6.0 ±0.6 8.8 ±0.6 7.1 ±0.2 8.7 ±1.1 6.9±0.0 7.3 ±0.2
Valine 5.7 ±0.1 6.0 ±0.2 6.0 ±0.5 6.2 ±0.0 5.7 ±0.8 5.9 ±0.6 5.9 ±0.0 6.2 ±0.1
Methionine N.D. 0.5 ±0.0 1.0 ±0.2 0.7 ±0.5 2.8 ±0.3 0.9 ±0.1 2.8 ±0.0 1.9 ±0.2
Isoleucine 2.4 ±0.1 4.7 ±0.2 3.9 ±0.4 3.8 ±0.6 4.3 ±0.1 4.0 ±0.7 4.9±0.1 4.0 ±0.1
Leucine 10.9 ±0.1 8.8 ±0.3 8.5 ±0.9 8.4 ±1.1 8.3 ±0.1 7.9 ±0.7 7.9±0.3 8.4 ±0.1
Tyrosine 3.8 ±0.1 4.1 ±0.1 3.8 ±0.4 2.1 ±0.3 3.2 ±0.1 3.3 ±0.7 5.5±0.2 3.8 ±0.3
Phenylalanine 5.8 ±0.1 4.8 ±0.2 6.4 ±0.6 5.0 ±0.7 5.6 ±0.2 5.3 ±0.1 5.6 ±0.4 5.4 ±0.1
Histidine 3.5 ±0.6 4.0 ±0.8 2.2 ±0.7 3.3 ±0.2 2.1 ±1.2 2.5 ±0.5 1.9±0.1 3.0 ±0.4
Lisine 11.4 ±0.1 7.4 ±0.2 6.9 ±0.6 6.3 ±0.1 5.5 ±0.3 5.2 ±0.4 5.3±0.0 7.1 ±0.3
Arginine 5.1 ±0.5 3.7 ±0.7 6.4 ±0.9 5.0 ±0.4 5.6 ±0.6 5.7 ±0.9 4.0±0.3 6.5 ±0.2
Ammonia 1.0 ±0.2 1.5 ±0.2 1.0 ±0.2 1.6 ±0.1 2.5 ±0.1 1.8 ±0.1 1.8±0.0 0.6 ±0.0
Total 99.8 ±3.0 96.5 ±6.7 94.8 ±5.5 96.4 ±4.7 94.9 ±5.5 98.7 ±2.3 95.2 ±2.5 98.8 ±3.6
Heterokontophyta Prymnesiophyta Rhodophyta
Amino acid C. minima D. menstrualis P. gymnospora S. costatum S. vulgare I. galbana A. uruguayense P. acanthophora P. capillacea
Cisteic acid 0.5 ±0.1 0.5 ±0.0 0.9 ±0.3 0.3 ±0.0 0.6 ±0.2 0.6 ±0.0 0.9 ±0.3 1.2 ±0.3 0.7 ±0.1
Aspartic acid 12.0 ±0.8 14.5 ±0.7 12.8 ±2.5 13.4 ±0.1 10.6 ±1.3 12.6 ±0.7 13.2 ±1.8 12.5 ±2.1 11.6 ±2.8
Treonine 5.1 ±0.3 5.0 ±0.0 5.1 ±0.6 5.2 ±0.1 4.4 ±0.7 5.1 ±0.4 5.4±0.6 5.8 ±0.3 5.2 ±0.9
Serine 6.0 ±0.5 6.8 ±0.4 5.0 ±0.6 4.7 ±0.1 4.7 ±0.7 4.1 ±0.4 5.2 ±0.2 5.3 ±0.4 5.7 ±1.4
Glutamic acid 14.8 ±1.4 12.6 ±0.4 13.1 ±1.4 13.5 ±0.0 17.4 ±0.4 12.1 ±0.2 14.9 ±1.5 12.9 ±4.3 14.7 ±0.2
Proline 4.3 ±0.4 4.8 ±0.1 4.3 ±0.7 3.7 ±0.0 4.2 ±0.7 4.1 ±0.5 4.9±0.4 4.6 ±0.1 4.9 ±0.6
Glycine 6.0 ±0.4 6.0 ±0.0 6.0 ±0.9 6.2 ±0.1 5.3 ±0.9 5.8 ±0.2 6.5±0.1 7.1 ±1.4 6.0 ±0.9
Alanine 7.9 ±0.8 6.6 ±0.2 6.9 ±0.5 6.7 ±0.1 6.8 ±1.1 7.4 ±0.3 7.5±0.7 8.8 ±1.2 7.2 ±1.4
Valine 5.7 ±0.4 5.2 ±0.1 5.3 ±0.6 5.9 ±0.0 5.4 ±0.9 6.4 ±0.2 6.0 ±0.7 6.4 ±0.2 5.5 ±1.8
Methionine 2.0 ±0.3 1.3 ±0.2 1.0 ±0.4 2.6 ±0.1 1.7 ±0.3 2.6 ±0.1 0.7±0.3 1.1 ±0.1 1.1 ±0.1
Isoleucine 3.9 ±0.4 4.3 ±0.0 4.3 ±0.3 5.7 ±0.0 4.3 ±0.8 5.1 ±0.2 4.7±0.2 4.1 ±0.8 3.7 ±0.6
Leucine 7.9 ±0.6 8.6 ±0.1 8.5 ±1.1 8.3 ±0.1 8.2 ±1.4 9.3 ±0.3 8.2±1.2 8.1 ±0.9 6.8 ±1.3
Tyrosine 1.8 ±0.3 2.6 ±0.1 2.1 ±0.4 3.2 ±0.1 1.8 ±0.2 3.4 ±0.2 2.4±0.3 2.4 ±0.4 3.7 ±0.4
Phenylalanine 4.9 ±0.3 5.5 ±0.1 5.2 ±0.7 6.1 ±0.1 4.9 ±0.8 5.9 ±0.1 5.2 ±0.6 4.7 ±1.0 5.3 ±0.9
Histidine 2.0 ±0.2 2.2 ±0.1 2.1 ±0.5 1.6 ±0.1 1.6 ±0.3 2.0 ±0.2 2.4±0.5 3.0 ±0.6 3.5 ±0.5
Lisine 5.0 ±0.4 4.6 ±0.2 5.4 ±0.8 4.6 ±1.2 5.0 ±0.9 5.4 ±0.4 6.2 ±0.9 6.3 ±1.4 7.9 ±0.8
Arginine 4.2 ±0.3 5.1 ±0.2 5.0 ±0.6 4.1 ±0.0 3.9 ±0.6 5.8 ±0.9 4.7±0.2 4.8 ±0.1 5.6 ±0.6
Ammonia 1.4 ±0.2 1.2 ±0.1 1.5 ±0.2 2.4 ±0.0 1.3 ±0.1 1.9 ±0.1 1.8±0.3 1.9 ±0.1 1.7 ±0.56
Total 95.2 ±4.7 98.0 ±2.6 94.5 ±6.4 95.8 ±2.3 94.0 ±4.6 98.3 ±3.2 96.6 ±4.6 99.2 ±6.1 99.0 ±5.3
aResults are expressed as percentage of amino acid per 100 g of algal protein (or pure protein for the two standards) and represent the real
recovery of amino acids after analysis. Concentrations of ammonia correspond to nitrogen recovery from some amino acids destroyed during
acid hydrolysis. Values indicate the mean of three replicates ±S.D. (n(3). N.D.: not detected.
456
Table 3. Total nitrogen, total amino acid residues and protein content of marine algae, as percentage of the dry mattera
Lowry precipitation Bradford precipitation Lowry row extract,
with TCA 2.5:1 with TCA 2.5:1 no precipitation
Total Total amino Crude
Species nitrogen BSA Casein BSA Casein BSA Casein acid residues protein
Chlorophyta
C. fastigiata 4.32 ±0.36 5.47 ±0.34 5.29 ±0.33 1.74 ±0.08 1.63 ±0.07 7.52 ±0.58 7.27 ±0.56 13.50 ±2.31 19.53
C. decorticatum 2.13 ±0.10 7.12 ±0.53 6.88 ±0.51 4.49 ±0.37 4.24 ±0.35 7.55 ±0.08 7.30 ±0.08 10.93 ±1.06 11.37
D. tertiolecta 4.18 ±0.09 11.4 ±0.99 11.0 ±0.95 6.86 ±0.26 6.48 ±0.25 11.0 ±0.13 10.6 ±0.12 17.14 ±0.70 18.35
U. fasciata 2.29 ±0.02 7.30 ±0.84 7.05 ±0.81 2.60 ±0.19 2.43 ±0.23 7.55 ±0.10 7.37 ±0.09 11.03 ±0.98 12.80
Cryptophyta
Hillea sp. 5.08 ±0.26 15.3 ±0.60 14.8 ±0.58 8.54 ±0.35 8.07 ±0.33 13.1 ±0.51 12.7 ±0.49 20.11 ±2.22 25.04
Dinophyta
A. carterae 4.69 ±0.05 10.2 ±0.09 9.85 ±0.09 5.49 ±0.26 5.18 ±0.25 10.8 ±0.12 10.5 ±0.12 15.84 ±1.72 24.06
Heterokontophyta
C. minima 1.94 ±0.10 7.83 ±0.33 7.57 ±0.32 3.56 ±0.23 3.36 ±0.22 10.0 ±0.20 9.67 ±0.20 9.99 ±0.80 11.06
D. menstrualis 3.26 ±0.15 4.04 ±0.17 3.90 ±0.16 2.72 ±0.21 2.56 ±0.20 7.01 ±0.13 6.77 ±0.13 10.35 ±0.31 14.83
P. gymnospora 2.41 ±0.14 8.69 ±0.72 8.40 ±0.70 4.79 ±0.45 4.53 ±0.43 11.9 ±0.50 11.5 ±0.48 12.55 ±1.55 13.78
S. costatum 3.41 ±0.22 11.1 ±0.68 10.7 ±0.66 6.43 ±0.39 6.07 ±0.37 11.5 ±0.75 11.1 ±0.72 14.30 ±1.76 15.40
S. vulgare 2.08 ±0.14 6.91 ±0.15 6.68 ±0.14 3.19 ±0.19 3.00 ±0.18 8.77 ±0.12 8.47 ±0.12 11.0 ±1.54 11.50
Prymnesiophyta
I. galbana 3.35 ±0.20 10.1 ±0.85 9.75 ±0.82 6.58 ±0.49 6.22 ±0.46 11.1 ±0.26 10.7 ±0.25 15.83 ±0.09 16.98
Rhodophyta
A. uruguayense 5.68 ±0.03 15.7 ±0.33 15.1 ±0.32 10.2 ±0.32 9.48 ±0.30 12.2 ±0.22 11.7 ±0.21 17.22 ±1.88 22.38
P. acanthophora 3.68 ±0.04 4.19 ±0.29 4.05 ±0.28 2.60 ±0.13 2.45 ±0.12 6.26 ±0.18 6.05 ±0.17 11.83 ±1.58 16.45
P. capillacea 3.24 ±0.10 8.94 ±0.53 7.98 ±0.49 4.62 ±0.18 4.23 ±0.19 11.7 ±0.38 10.9 ±0.35 12.11 ±3.00 15.49
aProtein was determined by different methods, with BSA and casein as protein standards. Analysis by Lowry’s method was also made with
notprecipitated samples. Data represent the mean of six replicates ±S.D. (n(6), except for total nitrogen and amino acid residues (n(3).
a greater volume of water seems to improve the ex-
traction of protein. Differences in the behaviour of
the two flattened algae may be related to the chemi-
cal composition and thallus morphology, since U. fas-
ciata has two layers of cells compared to P. acan-
thophora var. acanthophora with only one layer of
cells.
For the extraction of protein from the branched and
hard thallus of the brown algae S. vulgare,12hin-
cubation in water seems to be important. Even us-
ing a greater volume of water (4mL), an incubation
period of 6 h was not enough to soften the thalli.
Fragments of S. vulgare thallus were ground more eas-
ily with the Potter homogeniser after 12 h of exposure to
water.
As the precipitation of protein with TCA: ho-
mogenate (3.0:1 and 2.5:1 v/v) gave no significantly
different results and therefore a TCA:homogenate
ratio of 2.5:1 (v/v) is recommended in order to save
reagent.
Amino acid profile, non-protein nitrogen and protein
content of algae
In this study we assume that the actual concentra-
tions of protein in the samples are calculated from
the sum of amino acid residues (Tables 1 and 2), a
widely accepted procedure since the 1970s (Heidel-
baugh et al.,1975). After acid hydrolysis, all proteins
are destroyed, even those associated with other macro-
molecules and biological membranes. The values for
the total amino acid residues were calculated by sum-
ming up the amino acid masses retrieved after acid hy-
drolysis (total amino acid), less the water mass (18 g in
1Mof each amino acid) incorporated into each amino
acid after disruption of the peptide bonds. Total amino
acid analysis involves some errors, such as the total
(tryptophan) or partial (methionine and cysteine) de-
struction of some amino acids, as well as the impossi-
bility of identifying the contribution of free amino acids
in the samples. However, it indicates the maximum
457
possible concentration of protein in the sample con-
sidering that all amino acid are in protein, providing
a good reference point for the protein concentrations
measured by the Bradford and Lowry assays. In our
results, values for amino acid residues were similar to
the estimated crude protein, suggesting the suitability
of the nitrogen-to-protein conversion factors calculated
by Louren¸co et al. (2002, 2004). Exceptions to this are
C. fastigiata and A. carterae (Table 3), which showed
values for crude protein ca. 30% higher than those for
the sum of amino acid residues. This difference prob-
ably results from the presence of high concentrations
of non-protein nitrogen, presumably transient stocks
of inorganic nitrogen (Lav´ın & Louren¸co, unpublished
data).
The sum of amino acid residues indicates that mi-
croalgae, independent of the taxonomic group, tend
to accumulate higher concentrations of protein than
macroalgae. This fact may be related to the high con-
centration of nitrogen in the culture medium (as well as
other dissolved nutrients), growth conditions in the lab-
oratory and the higher surface area:volume ratios found
in microalgae (Hein et al., 1995). Dried samples of mi-
croalgae are better exposed to solvents and to grind-
ing during the extraction procedures, while macroalgal
samples have to be powdered before the start of the
extraction. This factor may produce a more efficient
protein extraction.
For many macroalgae, the combined concentration
of glutamic acid and aspartic acid represents 40% of
total amino acids, agreeing with data obtained for the
edible red algae Palmaria palmata,inwhich glu and
asp represent 39.6% of total amino acids (Galland-
Irmouli et al.,1999). For microalgae, the sum of asp
and glu represent mean values of about 20% of the total
amino acids for Skeletonema costatum,Dunaliella ter-
tiolecta and Thalassiosira pseudonana (Brown, 1991).
For Ulva rigida and U. rotundata, percentages of these
two amino acid may represent from 26 to 32% of
the total amino acids (Fleurence et al., 1995). In the
present study, values for asp +glu varied from 19.2%
(C. fastigiata)to28.1% (A. uruguayense)(Table 2).
For microalgae, the fraction represented by these two
amino acids varied from 23.1% (A. carterae)to26.9%
(S. costatum)ofthe total amino acids (Table 2). The
set composed of the essential amino acids in samples
varied from 36.3% (S. vulgare)to44.0% (C. fasti-
giata), with a mean value of 40.2% of the total amino
acids. Concerning nutritional properties, these species
show concentrations of essential amino acids compa-
rable to those commonly described to soybean pro-
tein, which possesses 36.0% of the total amino acid
(Galland-Irmouli et al., 1999).
High concentrations of non-protein N may result
in overestimation of protein (Zamer et al., 1989). Ac-
cording to Louren¸co et al. (2004), concentrations of
non-protein N vary widely during growth in cultures
of microalgae, commonly fluctuating from 15 to 30%
of the total N. In the present study, the occurrence
of high concentrations of the total N was not mir-
rored by high total amino acid concentrations in some
species such as A. carterae,A. uruguayense,C. fasti-
giata, and Hillea sp., which is explained by the pres-
ence of large amounts on non-protein N. The deter-
mination of crude protein should be based on the use
of specific nitrogen-to-protein conversion factors as
proposed by Louren¸co et al. (2002, 2004). In addi-
tion, results obtained for the total amino acid residues
and precipitated and non-precipitated extracts with the
Lowry method suggest the influence of variable con-
centrations of free amino acids and small peptides in
the samples. This finding indicates that precipitation
of the samples is a fundamental step during protein
analysis.
Data of protein content in macroalgae from the trop-
ical and subtropical coastal environments frequently
show lower concentrations (Kaehler & Kennish, 1996;
Wong & Cheung, 2000). In some Brazilian environ-
ments Ramos et al. (2000) found that the percentage
of protein (N ×6.25) in 14 seaweeds varied from 2.30
to 25.6% of dry weight. Despite the overestimation of
protein content caused by the use of the factor 6.25
(Louren¸co et al., 2002), values obtained by Ramos
et al. (2000) indicated predominantly low concentra-
tions of protein. This trend may be related to the nat-
ural characteristics of Brazilian marine environments;
predominantly oligotrophic, with low availability of N
(Oliveira et al., 1997; Ovalle et al., 1999). As a conse-
quence, low concentrations of protein would be accu-
mulated by natural populations of macroalgae. In this
context, our data of protein concentration in macroal-
gae are in accordance with the information available
in the literature (e.g. Wong & Cheung, 2001; McDer-
mid & Stuercke, 2003). On the other hand, the relative
low content of protein in microalgae results from the
physiological state of the species. All microalgae were
sampled in stationary growth phase when percentages
of protein in cells decreased due to depletion of dis-
solved nutrients in the culture medium (Louren¸co et al.,
1998).
Low protein levels were found in Pterocladiella
capillacea using both the Lowry and Bradford
458
methods. The values obtained are equivalent to 1/3
of those determined from the sum of the amino
acid residues, being the lowest protein concentrations
among all red macroalgae tested. This seems to indicate
inefficient extraction using the procedures developed in
this study for this macroalgae. We hypothesise that the
extraction of protein in this species might be influenced
by the presence of phycocolloids, especially agarans.
P. capillacea is a good source of agar (McHugh, 1991)
and is probably the most abundant macromolecule in
this alga. During the water extraction step, it was possi-
ble to extract variable quantities of agar visible, as the
occurrence of gels in several steps of protein extrac-
tion, mainly when using the Potter homogeniser and
after the centrifugation at 4 ◦Cofthe ground samples.
It is possible that the gels can trap part of the protein
extracted, giving low values in the spectrophotometric
determination of protein. This kind of analytical prob-
lem may be present in other agarophytes as well as
carragheenan-producing species. The presence of large
amounts of anionic polysaccharide in the cell walls re-
duces protein solubility during extraction (Fleurence,
1999). Further studies are needed to develop better
procedures for protein extraction in phycocolloid-rich
species.
Lowry ×Bradford methods and the influence of the
amino acid profile of samples and standards
Some authors suggest that Bradford’s method would
generate lower protein values for a large number of or-
ganisms compared to Lowry’s method. Calculating the
Lowry:Bradford ratio from data by Eze & Dumbroff
(1982) for leaves of bean plants gives a ratio of 1.4. For
the diatom T. pseudonana, Clayton et al. (1988) found
ratios varying from 1.8 to 2.0, while Berges et al. (1993)
determined a ratio of 1.2 for the same microalgae. The
present results confirm the general trends found by
those authors, but we found higher Lowry:Bradford
ratios for most of the species. Our results varied from
1.5 (I. galbana, BSA as the protein standard) to 3.2 (C.
fastigiata, casein as the protein standard).
The trend of obtaining lower concentrations of
protein using Bradford’s method may be related to the
binding of the dye Coomassie Brilliant Blue-G250 to
both basic and aromatic amino acid residues (Compton
&Jones, 1985). Most of the algae show relatively
low concentrations of the two amino acids (tyrosine
and tryptophan) as well as the two basic amino acids
(lysine and histidine). Thus, the binding of the dye
with protein occurs mainly with the two amino acids,
arginine and phenylalanine, and this fact seems to
contribute to lower protein measurements. Our results
with Bradford’s method agree with Kaehler and
Kennish (1996). These authors found predominantly
low values for some seaweeds (from 1.3 to 12.6%)
from Hong Kong using the Bradford method. In
contrast, the Folin–Ciocalteu reagent used in the
Lowry assay interacts with all peptide bonds and also
with some amino acids. As a result, the quantification
of protein tends to be greater.
The differences in amino acid composition among
protein standards and algae have important implica-
tions regarding protein reactivity and quantification in
algal samples. Despite the good linearity obtained with
both protein standards (BSA and casein), our data sug-
gest that casein has a slightly smaller reactivity than
BSA resulting in a smaller quantification of protein.
The two protein standards have extremely different
amino acid composition and the reactivity of them in
each method tends to be different due the functional
groups that they present (Morrison & Boyd, 2003).
Functional groups change the charge and the geom-
etry of neighbouring atoms, affecting the reactivity of
the whole molecule (Morrison & Boyd, 2003).
Conclusions
The use of 4.0 mL of water to incubate algal samples for
12 h, combined with grinding of the samples with a Pot-
ter homogeniser is strongly recommended. This proce-
dure results in better extraction of protein from algal
samples of different species independent of their mor-
phological and biochemical characteristics. As a conse-
quence, this procedure can be applied widely to many
algal species. The precipitation of protein should be
done with 25% trichloroacetic acid in the ratio of 2.5:1
(TCA:homogenate). Results generated with Lowry’s
method are more similar to the data obtained from the
sum of the amino acid residues which is considered
the most reliable way of determining the actual pro-
tein content. Protein values obtained with BSA as a
protein standard were closer to those calculated from
the sum of amino acid residues, suggesting that the
use of BSA is more suitable for the Lowry method.
The procedures proposed here can contribute to better
results, since protein is extracted efficiently and po-
tential interference from compounds such as pigments,
lipids, phenolics, small peptides and free amino acids is
eliminated.
459
Acknowledgments
We are indebted to FAPERJ (Foundation for Re-
search Support of Rio de Janeiro State, grant E-
26.170.041/98) for the financial support to this study.
Special acknowledgements are due to Dr. Yocie
Yoneshigue-Valentin (Universidade Federal do Rio de
Janeiro) and Dr. Carlos Logullo de Oliveira (Uni-
versidade Estadual do Norte Fluminense) for offer-
ing us laboratory facilities to perform this study. We
thank Dr Ursula M. Lanfer Marquez (Universidade de
S˜ao Paulo) for her support in the amino acid analy-
sis. E.B. acknowledges CAPES and S.O.L. acknowl-
edges FAPERJ and CNPq for providing them research
fellowships.
References
Berges JA, Fischer AE, Harrison PJ (1993) A comparison
of Lowry, Bradford and Smith protein assays using dif-
ferent protein standards and protein isolated from marine
diatom Thalassiosira pseudonana. Mar. Biol. 115: 187–
193.
Bradford MM (1976) A rapid and sensitive method for the quantita-
tion of microgram quantities of protein utilizing the principle of
protein-dye binding. Anal. Biochem. 72: 248–254.
Brown MR (1991) The amino-acid and sugar composition of 16
species of microalgae used in mariculture. J. Exp. Mar. Biol.
Ecol. 145: 79–99.
Clayton JR Jr, Dortch Q, Thoresen SS, Ahmed SI (1988) Evaluation
of methods for the separation and analysis of proteins and free
amino acids in phytoplankton samples. J. Plankton Res. 10: 341–
358.
Compton SJ, Jones CG (1985) Mechanism of dye response and in-
terference in the Bradford protein assay. Anal. Biochem. 151:
369–374.
Conklin-Brittain NL, Dierenfeld ES, Wrangham RW, Norconk M,
Silver SC (1999) Chemical protein analysis: A comparison of
Kjeldahl crude protein and total ninhydrin protein from wild,
tropical vegetation. J. Chem. Ecol. 25: 2601–2622.
Crossman DJ, Clements KD, Cooper GJS (2000) Determination of
protein for studies of marine herbivory: A comparison of meth-
ods. J. Exp. Mar. Biol. Ecol. 244: 45–65.
Dortch Q, Clayton JR Jr, Thoresen SS, Ahmed SI (1984) Species
differences in accumulation of nitrogen pools in phytoplankton.
Mar. Biol. 81: 237–250.
Eze JMO, Dumbroff EB (1982) A comparison of the Bradford and
Lowry methods for the analysis of protein in chlorophyllous tis-
sue. Can. J. Bot. 60: 1046–1049.
Ezeagu IE, Petzke JK, Metges CC, Akinsoyinu AO, Ologhobo AD
(2002) Seed protein contents and nitrogen-to-protein conversion
factors for some uncultivated tropical plant seeds. Food Chem.
78: 105–109.
Fleurence J (1999) The enzymatic degradation of algal cell walls:
A useful approach for improving protein accessibility? J. Appl.
Phycol. 11: 313–314.
Fleurence J, Le Couer C, Mabeau S, Maurice M, Landrein A (1995)
Comparison of different extractive procedures for proteins from
the edible seaweeds Ulva rigida and Ulva rotundata.J.Appl.
Phycol. 7: 577–582.
Folin O, Ciocalteu V (1927) On tyrosine and tryptophane determi-
nations in proteins. J. Biol. Chem. 73: 627–650.
Fujihara S, Kasuga A, Aoyagi Y (2001) Nitrogen-to-protein conver-
sion factors for common vegetables in Japan. J. Food Sci. 66:
412–415.
Galland-Irmouli A-V, Fleurence J, Lamghari R, Lu¸con M, Rouxel
C, Barbaroux O, Bronowicki J-P, Villaume C, Gu´eant J-L (1999)
Nutritional value of proteins from edible seaweed Palmaria pal-
mata (Dulse). J. Nutr. Biochem. 10: 353–359.
Heidelbaugh ND, Huber CS, Bednarczyk JF, Smith MC, Rambaut
PC, Wheeler HO (1975) Comparison of three methods for
calculating protein content of foods. J. Agric. Food Chem. 23:
611–613.
Hein M, Pedersen MF, Sand-Jensen K (1995) Size-dependent
nitrogen uptake in micro- and macroalgae. Mar. Ecol. Prog. Ser.
118: 247–253.
Kaehler S, Kennish R (1996) Summer and winter comparisons in
the nutritional value of marine macroalgae. Hong Kong Bot.
Mar. 39: 11–17.
Kester DR, Duedale IW, Connors DN, Pytkowicz RM (1967)
Preparation of artificial sea water. Limnol Oceanogr. 12:
176–179.
Lee RE (1999) Phycology. 3rd ed., Cambridge University Press,
Cambridge, 614 pp.
Legler G, M¨uller-Platz CM, Mentges-Hettkamp M, Pflieger G,
J¨ulich E (1985) On the chemical basis of the Lowry protein
determination. Anal. Biochem. 150: 278–287.
Louren¸co SO, Barbarino E, De-Paula JC, Pereira LOS, Lanfer
Marquez UM (2002) Amino acid composition, protein content,
and calculation of nitrogen-to-protein conversion factors for
nineteen tropical seaweeds. Phycol. Res. 50: 233–241.
Louren¸co SO, Barbarino E, Lanfer Marquez UM, Aidar E (1998)
Distribution of intracellular nitrogen in marine microalgae: Basis
for the calculation of specific nitrogen-to-protein conversion
factors. J. Phycol. 34: 798–811.
Louren¸co SO, Barbarino E, Lav´ın PL, Lanfer Marquez UM,
Aidar E (2004) Distribution of intracellular nitrogen in marine
microalgae. Calculation of new nitrogen-to-protein conversion
factors. Eur. J. Phycol. 39: 17–32.
Lowry OH, Rosebrough NJ, Farr AL, Randall RJ (1951) Protein
measurement with the folin phenol reagent. J. Biol. Chem. 193:
265–275.
Mattoo RL, Ishaq M, Saleemuddin M (1987) Protein assay by
Coomassie Brilliant Blue G-250-binding method is unsuitable
for plant tissues rich in phenols and phenolases. Anal. Biochem.
163: 376–384.
McDermid KJ, Stuercke B (2003) Nutritional composition of edible
Hawaiian seaweeds. J. Appl. Phycol. 15: 513–524.
McHugh DJ (1991) Worldwide distribution of commercial re-
sources of seaweeds including Gelidium. Hydrobiologia 221:
19–29.
Morrison RT, Boyd RN (2003) Organic Chemistry, 7th ed., Allyn
and Bacon, Boston, 1200 pp.
Moss´eJ(1990) Nitrogen to protein conversion factor for ten cereals
and six legumes or oilseeds. A reappraisal of its definition and
determination. Variation according to species and to seed proteic
content. J. Agric. Food Chem. 38: 18–24.
460
Nguyen RT, Harvey HR (1994) A rapid micro-scale method for
the extraction and analysis of protein in marine samples. Mar.
Chem. 45: 1–14.
Oliveira EC, Corbisier TN, de Eston VR, Ambrosio O (1997) Phe-
nology of a seagrass (Halodule wrightii) bed on the southeast
coast of Brazil. Aqua. Bot. 56: 25–33.
Ovalle ARC, Rezende CE, Carvalho CEV, Jennerjahn TC, Ittekkot
V (1999) Biogeochemical characteristics of coastal waters
adjacent to small river-mangrove systems, East Brazil Geo-Mar.
Lett. 19: 179–185.
P´erez-Llor´ens JL, Ben´ıtez E, Vergara JJ, Berges JA (2003) Charac-
terization of proteolytic enzyme activities in macroalgae. Eur. J.
Phycol. 38: 55–64
Peterson GL (1979) Review of the folin phenol protein quantita-
tion method of Lowry, Rosebrough, Farr and Randall. Anal.
Biochem. 100: 201–220.
Ramos MV, Monteiro ACO, Moreira RA, Carvalho AFFU (2000)
Amino acid composition of some Brazilian seaweed species. J.
Food Biochem. 24: 33–39.
Stoscheck CM (1990) Quantitation of protein. Methods Enzymol.
182: 50–68.
Walne PR (1966) Experiments in the large scale culture of the
larvae of Ostrea edulis. Fishery Invest. (Lond.) (Ser 2) 25(4):
1–53.
Wong KH, Cheung CK (2000) Nutritional evaluation of some sub-
tropical red and green seaweeds Part I—proximate composition,
amino acid profiles and some physico-chemical properties. Food
Chem. 71: 475–482.
Wong KH, Cheung CK (2001) Nutritional evaluation of some
subtropical red and green seaweeds Part II. In vitro protein
digestibility and amino acid profiles of protein concentrates.
Food Chem. 72: 11–17.
Wynne MJ (1998) A checklist of benthic marine algae of the tropical
and subtropical western Atlantic: First revision. Nova Hedwigia
116: 1–155.
Zamer WE, Shick JM, Tapley DW (1989) Protein measurement
and energetic considerations: Comparisons of biochemical
and stoichiometric methods using bovine serum albumin and
protein isolated from sea anemones. Limnol. Oceanogr. 34: 256–
263.
Zar JH (1996) Biostatistical Analysis. 3rd ed., Prentice Hall, Inc.,
Upper Saddle River, USA. 920 pp.